
#!pip install folium


#!pip install imbalanced-learn


#Making necessary imports
import numpy as np  # linear algebra
import pandas as pd # data processing
import plotly.express as px
import matplotlib.pyplot as plt
import seaborn as sns
import os, sys
import folium
import sklearn
from sklearn import svm
from sklearn.svm import SVC
from sklearn import metrics
from sklearn import tree
from sklearn.metrics import classification_report, confusion_matrix
from sklearn.metrics import accuracy_score
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import cross_val_score, StratifiedKFold
from sklearn.neighbors import NeighborhoodComponentsAnalysis,KNeighborsClassifier
from sklearn.model_selection import GridSearchCV, RandomizedSearchCV
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.metrics import precision_score, recall_score , f1_score
from sklearn.feature_selection import SelectKBest, f_classif
from sklearn.metrics import roc_curve, roc_auc_score
from sklearn.naive_bayes import GaussianNB
from imblearn.over_sampling import SMOTE
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from scipy.stats import chi2_contingency
from sklearn.utils import shuffle
from sklearn.preprocessing import MinMaxScaler
from statsmodels.stats.outliers_influence import variance_inflation_factor
from sklearn.preprocessing import StandardScaler
import warnings
from matplotlib.colors import LinearSegmentedColormap
from sklearn.metrics import precision_score, recall_score, ConfusionMatrixDisplay
from itertools import cycle
from sklearn.tree import plot_tree
from scipy.stats import randint
from sklearn import model_selection
from sklearn.pipeline import make_pipeline
from sklearn.metrics import precision_recall_curve
import warnings
from sklearn.exceptions import ConvergenceWarning


# Loading in the Dataset
Churn =pd.read_csv("Churn.csv")


#Checking head of dataset (first 10)
Churn.head(n=10)


# Checking size of dataset
Churn.shape

(7043, 38)


# Checking summary statistics of numerical features
Churn.describe()


#Checking data types of each features and the number of non-null values.
Churn.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7043 entries, 0 to 7042
Data columns (total 38 columns):
 #   Column                             Non-Null Count  Dtype  
---  ------                             --------------  -----  
 0   Customer ID                        7043 non-null   object 
 1   Gender                             7043 non-null   object 
 2   Age                                7043 non-null   int64  
 3   Married                            7043 non-null   object 
 4   Number of Dependents               7043 non-null   int64  
 5   City                               7043 non-null   object 
 6   Zip Code                           7043 non-null   int64  
 7   Latitude                           7043 non-null   float64
 8   Longitude                          7043 non-null   float64
 9   Number of Referrals                7043 non-null   int64  
 10  Tenure in Months                   7043 non-null   int64  
 11  Offer                              3166 non-null   object 
 12  Phone Service                      7043 non-null   object 
 13  Avg Monthly Long Distance Charges  6361 non-null   float64
 14  Multiple Lines                     6361 non-null   object 
 15  Internet Service                   7043 non-null   object 
 16  Internet Type                      5517 non-null   object 
 17  Avg Monthly GB Download            5517 non-null   float64
 18  Online Security                    5517 non-null   object 
 19  Online Backup                      5517 non-null   object 
 20  Device Protection Plan             5517 non-null   object 
 21  Premium Tech Support               5517 non-null   object 
 22  Streaming TV                       5517 non-null   object 
 23  Streaming Movies                   5517 non-null   object 
 24  Streaming Music                    5517 non-null   object 
 25  Unlimited Data                     5517 non-null   object 
 26  Contract                           7043 non-null   object 
 27  Paperless Billing                  7043 non-null   object 
 28  Payment Method                     7043 non-null   object 
 29  Monthly Charge                     7043 non-null   float64
 30  Total Charges                      7043 non-null   float64
 31  Total Refunds                      7043 non-null   float64
 32  Total Extra Data Charges           7043 non-null   int64  
 33  Total Long Distance Charges        7043 non-null   float64
 34  Total Revenue                      7043 non-null   float64
 35  Customer Status                    7043 non-null   object 
 36  Churn Category                     1869 non-null   object 
 37  Churn Reason                       1869 non-null   object 
dtypes: float64(9), int64(6), object(23)
memory usage: 2.0+ MB


# Checking data types
Churn.dtypes

Customer ID                           object
Gender                                object
Age                                    int64
Married                               object
Number of Dependents                   int64
City                                  object
Zip Code                               int64
Latitude                             float64
Longitude                            float64
Number of Referrals                    int64
Tenure in Months                       int64
Offer                                 object
Phone Service                         object
Avg Monthly Long Distance Charges    float64
Multiple Lines                        object
Internet Service                      object
Internet Type                         object
Avg Monthly GB Download              float64
Online Security                       object
Online Backup                         object
Device Protection Plan                object
Premium Tech Support                  object
Streaming TV                          object
Streaming Movies                      object
Streaming Music                       object
Unlimited Data                        object
Contract                              object
Paperless Billing                     object
Payment Method                        object
Monthly Charge                       float64
Total Charges                        float64
Total Refunds                        float64
Total Extra Data Charges               int64
Total Long Distance Charges          float64
Total Revenue                        float64
Customer Status                       object
Churn Category                        object
Churn Reason                          object
dtype: object


Churn = Churn[Churn['Customer Status'] != 'Joined']


print(Churn['Customer Status'])

0        Stayed
1        Stayed
2       Churned
3       Churned
4       Churned
         ...   
7037    Churned
7038     Stayed
7039    Churned
7041     Stayed
7042     Stayed
Name: Customer Status, Length: 6589, dtype: object


# Checking shape of datset after removing joined
Churn.shape

(6589, 38)


#Checking tail of dataset (last 10)
Churn.tail(n=10)


categorical_columns = ['Gender','Married','City', 'Offer', 'Phone Service', 'Multiple Lines', 'Internet Service',
                       'Internet Type', 'Online Security', 'Number of Dependents', 'Number of Referrals', 'Online Backup',
                       'Device Protection Plan', 'Premium Tech Support', 'Streaming TV', 
                       'Streaming Movies', 'Streaming Music', 'Unlimited Data', 'Contract','Total Extra Data Charges', 'Paperless Billing', 'Payment Method', 'Churn Category', 'Churn Reason', 'Customer Status']
numerical_columns = ['Age', 'Tenure in Months', 'Avg Monthly Long Distance Charges',
                     'Avg Monthly GB Download', 'Monthly Charge', 'Total Charges', 'Total Refunds', 
                     'Total Long Distance Charges', 'Total Revenue']


Categorical = Churn[categorical_columns]
Numerical = Churn[numerical_columns]


#Checking head of datase
Categorical.head()


#Checking head of dataset 
Numerical.head()


# Defining the number of rows and columns for the grid
num_cols = 5
num_rows = (len(Categorical.columns) - 2 + num_cols - 1) // num_cols  
# Creating a grid of subplots
fig, axes = plt.subplots(num_rows, num_cols, figsize=(5*num_cols, 5*num_rows))
# Flatten the axes array
axes = axes.flatten()
# Setting color palette
color = sns.set_palette("pastel")
# Columns to iterate over excluding "city" and "churn reason"
columns_to_plot = [col for col in Categorical.columns if col not in ["City", "Churn Reason"]]
# Iterating over each column in the DataFrame
for i, column in enumerate(columns_to_plot):
    ax = axes[i]  # Get the subplot axes
    sns.countplot(x=column, data=Categorical, ax=ax, palette= color)
    ax.tick_params(axis='x', rotation=45)  # Rotate x-axis labels for better readability
# Remove empty subplots
for i in range(len(columns_to_plot), num_rows * num_cols):
    fig.delaxes(axes[i])
# Adjust layout
plt.tight_layout()
# Save image to include in PDF
plt.savefig('Distribution_of_Categorical_Feature.jpg')  
# Show the plot
plt.show()


# Setting color palette
sns.set_palette("husl", 9)
# Defining the number of rows and columns for the grid
num_cols = 5
num_rows = (len(Numerical.columns) + num_cols - 1) // num_cols
# Creating a grid of subplots
fig, axes = plt.subplots(num_rows, num_cols, figsize=(5*num_cols, 5*num_rows))
# Flatten the axes array
axes = axes.flatten()
# Columns to iterate over excluding "Total Refunds"
columns_to_plot = [col for col in Numerical.columns if col != "Total Refunds"]
# Iterating over each column in the DataFrame
for i, column in enumerate(columns_to_plot):
    ax = axes[i]  # Get the subplot axes
    # Plot boxplot for numerical columns
    sns.boxplot(x=column, data=Numerical, ax=ax)
    ax.set_title(f"Boxplot of {column}")
# Remove empty subplots
for i in range(len(columns_to_plot), num_rows * num_cols):
    fig.delaxes(axes[i])
# Adjust layout
plt.tight_layout()
# Save image to include in PDF
plt.savefig('Boxplot_of_each_Numerical_Feature.jpg')  
plt.show()


# Define the target variable
target_variable = 'Customer Status'
# Exclude columns from categorical_columns
columns_to_plot = [col for col in categorical_columns if col not in ['City', 'Churn Reason', 'Customer Status', 'Churn Category']]
# Calculate the number of columns for the subplot grid
num_cols = 3  # You may adjust this value as needed
# Calculate the actual number of rows needed
num_rows = (len(columns_to_plot) + num_cols - 1) // num_cols
# Create a grid of subplots
fig, axes = plt.subplots(num_rows, num_cols, figsize=(15, 5*num_rows))
# Flatten the axes array
axes = axes.flatten()
# Iterate over each categorical column
for i, column in enumerate(columns_to_plot):
    # Get the subplot axes
    ax = axes[i]    
    # Plot the relationship using a bar plot
    colour = ['pink','tan']
    sns.set_palette(colour)
    sns.countplot(x=column, hue=target_variable, data=Categorical, ax=ax)   
    # Set title
    ax.set_title(f'{column} vs {target_variable}')  
  # Rotate x-axis labels for better readability
    ax.tick_params(axis='x', rotation=45)
# Remove any remaining empty subplots
for j in range(len(columns_to_plot), num_rows * num_cols):
    fig.delaxes(axes[j])
# Adjust layout
plt.tight_layout()
plt.savefig("Bivariate Analysis.jpg")
plt.show()


# Group by city and churn status
city_churn_counts = Churn.groupby(['City', 'Customer Status']).size().unstack().reset_index()
city_churn_counts.fillna(0, inplace=True)  # Fill NaN values with 0 for cities with no churn data

# Create a map centered around a latitude and longitude
churn_map = folium.Map(location=[Churn['Latitude'].mean(), Churn['Longitude'].mean()], zoom_start=5)

# Add markers for each city
for index, row in city_churn_counts.iterrows():
    City = row['City']
    Churned = row['Churned'] if 'Churned' in row else 0
    Stayed = row['Stayed'] if 'Stayed' in row else 0
    marker_popup = f"{City}<br>Churned: {Churned}<br>Stayed: {Stayed}"
    folium.Marker(
        location=[Churn[Churn['City'] == City]['Latitude'].values[0], Churn[Churn['City'] == City]['Longitude'].values[0]],
        popup=marker_popup,
        icon=folium.Icon(color='red' if Churned > Stayed else 'green')
    ).add_to(churn_map)

# Add a tooltip to show city name and churned/stayed count when hovering over a marker
folium.Marker(
    location=[0, 0],  # Dummy location
    icon=None,
    tooltip=folium.Tooltip("<b>Click a marker to see details</b>")
).add_to(churn_map)

# Display the map
churn_map


#Removing unnecessary Columns
Churn.drop(["Customer ID", "Zip Code", "City", "Latitude", "Longitude", "Churn Reason", "Churn Category"], axis =1, inplace= True)


# Checking head of dataset after removing columns
Churn.head(n=30)


#Checking number of null in each column
Churn.isnull().sum()

Gender                                  0
Age                                     0
Married                                 0
Number of Dependents                    0
Number of Referrals                     0
Tenure in Months                        0
Offer                                3598
Phone Service                           0
Avg Monthly Long Distance Charges     644
Multiple Lines                        644
Internet Service                        0
Internet Type                        1344
Avg Monthly GB Download              1344
Online Security                      1344
Online Backup                        1344
Device Protection Plan               1344
Premium Tech Support                 1344
Streaming TV                         1344
Streaming Movies                     1344
Streaming Music                      1344
Unlimited Data                       1344
Contract                                0
Paperless Billing                       0
Payment Method                          0
Monthly Charge                          0
Total Charges                           0
Total Refunds                           0
Total Extra Data Charges                0
Total Long Distance Charges             0
Total Revenue                           0
Customer Status                         0
dtype: int64


# Calculate total number of rows
total_rows = Churn.shape[0]
# Calculate number of missing values in each column
missing_values = Churn.isnull().sum()
# Calculate percentage of missing values in each column
percentage_missing = (missing_values / total_rows) * 100
# Display the percentage of missing values in each column
print("Percentage of missing values in each column:")
print(percentage_missing)

Percentage of missing values in each column:
Gender                                0.000000
Age                                   0.000000
Married                               0.000000
Number of Dependents                  0.000000
Number of Referrals                   0.000000
Tenure in Months                      0.000000
Offer                                54.606162
Phone Service                         0.000000
Avg Monthly Long Distance Charges     9.773866
Multiple Lines                        9.773866
Internet Service                      0.000000
Internet Type                        20.397632
Avg Monthly GB Download              20.397632
Online Security                      20.397632
Online Backup                        20.397632
Device Protection Plan               20.397632
Premium Tech Support                 20.397632
Streaming TV                         20.397632
Streaming Movies                     20.397632
Streaming Music                      20.397632
Unlimited Data                       20.397632
Contract                              0.000000
Paperless Billing                     0.000000
Payment Method                        0.000000
Monthly Charge                        0.000000
Total Charges                         0.000000
Total Refunds                         0.000000
Total Extra Data Charges              0.000000
Total Long Distance Charges           0.000000
Total Revenue                         0.000000
Customer Status                       0.000000
dtype: float64


#Removing unnecessary Columns
Churn.drop(["Offer"], axis =1, inplace= True)


# Replacing missing values in numerical columns with the median
numerical_columns = ['Avg Monthly Long Distance Charges', 'Avg Monthly GB Download']
for column in numerical_columns:
    median_value = Churn[column].median()
    Churn[column].fillna(median_value, inplace=True)


# Replacing missing values in categorical columns with the mode
categorical_columns = ['Internet Type', 'Online Security', 'Multiple Lines', 'Internet Type', 'Online Security',
                       'Online Backup', 'Device Protection Plan', 'Premium Tech Support', 'Streaming TV', 'Streaming Movies',
                       'Streaming Music', 'Unlimited Data']  
for column in categorical_columns:
    mode_value = Churn[column].mode()[0]
    Churn[column].fillna(mode_value, inplace=True)


#Checking number of null in each column after replacing missing values
Churn.isnull().sum()

Gender                               0
Age                                  0
Married                              0
Number of Dependents                 0
Number of Referrals                  0
Tenure in Months                     0
Phone Service                        0
Avg Monthly Long Distance Charges    0
Multiple Lines                       0
Internet Service                     0
Internet Type                        0
Avg Monthly GB Download              0
Online Security                      0
Online Backup                        0
Device Protection Plan               0
Premium Tech Support                 0
Streaming TV                         0
Streaming Movies                     0
Streaming Music                      0
Unlimited Data                       0
Contract                             0
Paperless Billing                    0
Payment Method                       0
Monthly Charge                       0
Total Charges                        0
Total Refunds                        0
Total Extra Data Charges             0
Total Long Distance Charges          0
Total Revenue                        0
Customer Status                      0
dtype: int64


Churn.duplicated().sum()

0


# Checking head of dataset after replacing NaN Values
Churn.head(n=30)


# Checking data types before encoding
Churn.dtypes

Gender                                object
Age                                    int64
Married                               object
Number of Dependents                   int64
Number of Referrals                    int64
Tenure in Months                       int64
Phone Service                         object
Avg Monthly Long Distance Charges    float64
Multiple Lines                        object
Internet Service                      object
Internet Type                         object
Avg Monthly GB Download              float64
Online Security                       object
Online Backup                         object
Device Protection Plan                object
Premium Tech Support                  object
Streaming TV                          object
Streaming Movies                      object
Streaming Music                       object
Unlimited Data                        object
Contract                              object
Paperless Billing                     object
Payment Method                        object
Monthly Charge                       float64
Total Charges                        float64
Total Refunds                        float64
Total Extra Data Charges               int64
Total Long Distance Charges          float64
Total Revenue                        float64
Customer Status                       object
dtype: object


# Convert categorical columns (with 2 unique values) to numerical
# List of categorical columns excluding specified columns
categorical_columns = [col for col in Churn.select_dtypes(include=['object']).columns 
                       if col not in ['Internet Type', 'Contract', 'Payment Method']]
# Convert categorical columns to numerical using Label Encoding
for column in categorical_columns:
    le = LabelEncoder()
    Churn[column] = le.fit_transform(Churn[column])


# Checking head of dataset after Encoding features with 2 unique values
Churn.head()


# Extract the columns you want to encode
columns_to_encode = ['Internet Type', 'Contract', 'Payment Method']
# Apply One-Hot Encoding
encoded_columns = pd.get_dummies(Churn[columns_to_encode])
# Drop the original categorical columns from the DataFrame
Churn = Churn.drop(columns_to_encode, axis=1)
# Concatenate the encoded columns with the original DataFrame
Churn = pd.concat([Churn, encoded_columns], axis=1)


# Checking head of dataset after Encoding
Churn.head(n=10)


# Checking data types after encoding
Churn.dtypes

Gender                                 int32
Age                                    int64
Married                                int32
Number of Dependents                   int64
Number of Referrals                    int64
Tenure in Months                       int64
Phone Service                          int32
Avg Monthly Long Distance Charges    float64
Multiple Lines                         int32
Internet Service                       int32
Avg Monthly GB Download              float64
Online Security                        int32
Online Backup                          int32
Device Protection Plan                 int32
Premium Tech Support                   int32
Streaming TV                           int32
Streaming Movies                       int32
Streaming Music                        int32
Unlimited Data                         int32
Paperless Billing                      int32
Monthly Charge                       float64
Total Charges                        float64
Total Refunds                        float64
Total Extra Data Charges               int64
Total Long Distance Charges          float64
Total Revenue                        float64
Customer Status                        int32
Internet Type_Cable                     bool
Internet Type_DSL                       bool
Internet Type_Fiber Optic               bool
Contract_Month-to-Month                 bool
Contract_One Year                       bool
Contract_Two Year                       bool
Payment Method_Bank Withdrawal          bool
Payment Method_Credit Card              bool
Payment Method_Mailed Check             bool
dtype: object


# Select boolean columns
boolean_columns = Churn.select_dtypes(include='bool').columns
# Convert boolean columns to integers (0 and 1)
Churn[boolean_columns] = Churn[boolean_columns].astype(int)


# Checking head of dataset after Encoding
Churn.head(n=10)


# Select the features to be normalized
features_to_normalize = ['Avg Monthly Long Distance Charges', 'Tenure in Months', 'Age', 'Number of Referrals', 'Number of Dependents']

# Initialize the MinMaxScaler
scaler = MinMaxScaler()

# Apply Min-Max scaling to the selected features
Churn[features_to_normalize] = scaler.fit_transform(Churn[features_to_normalize])


# Checking head of dataset after normalisation
Churn.head(n=10)


# Correlation Analysis
# Selecting only numerical features (excluding the target variable)
numerical_features = Churn.select_dtypes(include='number').drop(columns=['Customer Status'])
# Compute correlation matrix
correlation_matrix = numerical_features.corr()
# Display correlation matrix
print(correlation_matrix)

                                     Gender       Age   Married  \
Gender                             1.000000  0.007664  0.004971   
Age                                0.007664  1.000000 -0.014409   
Married                            0.004971 -0.014409  1.000000   
Number of Dependents               0.006321 -0.125500  0.322628   
Number of Referrals               -0.004353 -0.032019  0.663598   
Tenure in Months                   0.011547 -0.009574  0.359737   
Phone Service                     -0.010518  0.007853  0.023813   
Avg Monthly Long Distance Charges  0.015720 -0.015403 -0.004581   
Multiple Lines                    -0.004748  0.098179  0.123751   
Internet Service                  -0.002584  0.110758 -0.021248   
Avg Monthly GB Download           -0.014879 -0.505260  0.066664   
Online Security                   -0.017059 -0.046073  0.132193   
Online Backup                     -0.009362  0.028884  0.125829   
Device Protection Plan            -0.001482  0.020439  0.136040   
Premium Tech Support              -0.012419 -0.052058  0.109226   
Streaming TV                      -0.001634 -0.037936  0.122905   
Streaming Movies                  -0.009462 -0.024728  0.117148   
Streaming Music                   -0.008470 -0.170316  0.071881   
Unlimited Data                     0.004778 -0.025313  0.029683   
Paperless Billing                 -0.010891  0.100351 -0.032990   
Monthly Charge                    -0.017288  0.124422  0.070009   
Total Charges                      0.004251  0.047788  0.296830   
Total Refunds                      0.008031  0.020570  0.031884   
Total Extra Data Charges          -0.000064  0.021759  0.012018   
Total Long Distance Charges        0.015564 -0.009976  0.243303   
Total Revenue                      0.008006  0.035080  0.308360   
Internet Type_Cable               -0.008537 -0.041248  0.006825   
Internet Type_DSL                  0.014655 -0.056854  0.003602   
Internet Type_Fiber Optic         -0.007227  0.078216 -0.007797   
Contract_Month-to-Month           -0.005772 -0.001283 -0.252296   
Contract_One Year                  0.008596 -0.010480  0.069478   
Contract_Two Year                 -0.001647  0.011245  0.214997   
Payment Method_Bank Withdrawal    -0.005451  0.114948 -0.003856   
Payment Method_Credit Card         0.007691 -0.101897  0.033121   
Payment Method_Mailed Check       -0.004661 -0.033586 -0.063845   

                                   Number of Dependents  Number of Referrals  \
Gender                                         0.006321            -0.004353   
Age                                           -0.125500            -0.032019   
Married                                        0.322628             0.663598   
Number of Dependents                           1.000000             0.276880   
Number of Referrals                            0.276880             1.000000   
Tenure in Months                               0.107311             0.323826   
Phone Service                                 -0.008652             0.012590   
Avg Monthly Long Distance Charges             -0.005649            -0.003455   
Multiple Lines                                -0.035975             0.067451   
Internet Service                              -0.164563            -0.051425   
Avg Monthly GB Download                        0.229396             0.064592   
Online Security                                0.036838             0.137056   
Online Backup                                 -0.005134             0.104668   
Device Protection Plan                        -0.024079             0.108669   
Premium Tech Support                           0.019176             0.104969   
Streaming TV                                   0.076270             0.110200   
Streaming Movies                               0.061325             0.089334   
Streaming Music                               -0.039533             0.042108   
Unlimited Data                                 0.029913             0.042395   
Paperless Billing                             -0.112755            -0.056960   
Monthly Charge                                -0.136297             0.011015   
Total Charges                                  0.016118             0.241943   
Total Refunds                                  0.012211             0.020372   
Total Extra Data Charges                      -0.015797            -0.002547   
Total Long Distance Charges                    0.065869             0.207028   
Total Revenue                                  0.032315             0.253851   
Internet Type_Cable                            0.006085             0.009476   
Internet Type_DSL                              0.010465             0.027511   
Internet Type_Fiber Optic                     -0.013380            -0.030775   
Contract_Month-to-Month                       -0.128389            -0.272251   
Contract_One Year                              0.004264             0.066514   
Contract_Two Year                              0.138544             0.239930   
Payment Method_Bank Withdrawal                -0.089444            -0.053171   
Payment Method_Credit Card                     0.099719             0.077497   
Payment Method_Mailed Check                   -0.018555            -0.050879   

                                   Tenure in Months  Phone Service  \
Gender                                     0.011547      -0.010518   
Age                                       -0.009574       0.007853   
Married                                    0.359737       0.023813   
Number of Dependents                       0.107311      -0.008652   
Number of Referrals                        0.323826       0.012590   
Tenure in Months                           1.000000       0.013282   
Phone Service                              0.013282       1.000000   
Avg Monthly Long Distance Charges          0.006230      -0.004927   
Multiple Lines                             0.299944       0.294161   
Internet Service                          -0.000288      -0.166607   
Avg Monthly GB Download                    0.034121      -0.032934   
Online Security                            0.315726      -0.097259   
Online Backup                              0.346104      -0.048827   
Device Protection Plan                     0.336471      -0.073776   
Premium Tech Support                       0.312314      -0.093205   
Streaming TV                               0.251734       0.113186   
Streaming Movies                           0.258661       0.103364   
Streaming Music                            0.211155      -0.040135   
Unlimited Data                            -0.003243       0.041242   
Paperless Billing                         -0.024360       0.012967   
Monthly Charge                             0.199019       0.243944   
Total Charges                              0.812795       0.123390   
Total Refunds                              0.042729       0.006675   
Total Extra Data Charges                   0.071224      -0.029174   
Total Long Distance Charges                0.653170       0.307688   
Total Revenue                              0.840935       0.189668   
Internet Type_Cable                       -0.005228      -0.224363   
Internet Type_DSL                          0.016244      -0.334472   
Internet Type_Fiber Optic                 -0.010868       0.447803   
Contract_Month-to-Month                   -0.614856      -0.002087   
Contract_One Year                          0.168878       0.000181   
Contract_Two Year                          0.524372       0.002147   
Payment Method_Bank Withdrawal            -0.025183       0.009661   
Payment Method_Credit Card                 0.084859       0.004319   
Payment Method_Mailed Check               -0.129430      -0.031003   

                                   Avg Monthly Long Distance Charges  \
Gender                                                      0.015720   
Age                                                        -0.015403   
Married                                                    -0.004581   
Number of Dependents                                       -0.005649   
Number of Referrals                                        -0.003455   
Tenure in Months                                            0.006230   
Phone Service                                              -0.004927   
Avg Monthly Long Distance Charges                           1.000000   
Multiple Lines                                             -0.005364   
Internet Service                                            0.011725   
Avg Monthly GB Download                                     0.013169   
Online Security                                             0.014592   
Online Backup                                               0.010468   
Device Protection Plan                                     -0.001754   
Premium Tech Support                                       -0.005476   
Streaming TV                                                0.000106   
Streaming Movies                                            0.009270   
Streaming Music                                             0.028029   
Unlimited Data                                             -0.012503   
Paperless Billing                                           0.015637   
Monthly Charge                                              0.012426   
Total Charges                                               0.011565   
Total Refunds                                              -0.024660   
Total Extra Data Charges                                    0.019051   
Total Long Distance Charges                                 0.548711   
Total Revenue                                               0.173532   
Internet Type_Cable                                         0.014270   
Internet Type_DSL                                          -0.008801   
Internet Type_Fiber Optic                                  -0.001829   
Contract_Month-to-Month                                    -0.002238   
Contract_One Year                                           0.030331   
Contract_Two Year                                          -0.025937   
Payment Method_Bank Withdrawal                              0.013207   
Payment Method_Credit Card                                 -0.010261   
Payment Method_Mailed Check                                -0.007024   

                                   Multiple Lines  Internet Service  ...  \
Gender                                  -0.004748         -0.002584  ...   
Age                                      0.098179          0.110758  ...   
Married                                  0.123751         -0.021248  ...   
Number of Dependents                    -0.035975         -0.164563  ...   
Number of Referrals                      0.067451         -0.051425  ...   
Tenure in Months                         0.299944         -0.000288  ...   
Phone Service                            0.294161         -0.166607  ...   
Avg Monthly Long Distance Charges       -0.005364          0.011725  ...   
Multiple Lines                           1.000000          0.199997  ...   
Internet Service                         0.199997          1.000000  ...   
Avg Monthly GB Download                 -0.001234          0.119774  ...   
Online Security                          0.083926          0.330946  ...   
Online Backup                            0.187246          0.380024  ...   
Device Protection Plan                   0.184029          0.381903  ...   
Premium Tech Support                     0.087188          0.333822  ...   
Streaming TV                             0.076823         -0.406993  ...   
Streaming Movies                         0.081905         -0.403762  ...   
Streaming Music                          0.180489          0.387690  ...   
Unlimited Data                          -0.037594         -0.177853  ...   
Paperless Billing                        0.154991          0.318405  ...   
Monthly Charge                           0.458214          0.721391  ...   
Total Charges                            0.451265          0.376125  ...   
Total Refunds                            0.036225          0.012650  ...   
Total Extra Data Charges                 0.057678          0.140662  ...   
Total Long Distance Charges              0.300490         -0.042382  ...   
Total Revenue                            0.448428          0.287064  ...   
Internet Type_Cable                     -0.101236          0.184681  ...   
Internet Type_DSL                       -0.149177          0.279211  ...   
Internet Type_Fiber Optic                0.200512         -0.372054  ...   
Contract_Month-to-Month                 -0.070235          0.223913  ...   
Contract_One Year                       -0.020749         -0.035474  ...   
Contract_Two Year                        0.097419         -0.215349  ...   
Payment Method_Bank Withdrawal           0.132185          0.259444  ...   
Payment Method_Credit Card              -0.093775         -0.225085  ...   
Payment Method_Mailed Check             -0.089814         -0.086532  ...   

                                   Total Revenue  Internet Type_Cable  \
Gender                                  0.008006            -0.008537   
Age                                     0.035080            -0.041248   
Married                                 0.308360             0.006825   
Number of Dependents                    0.032315             0.006085   
Number of Referrals                     0.253851             0.009476   
Tenure in Months                        0.840935            -0.005228   
Phone Service                           0.189668            -0.224363   
Avg Monthly Long Distance Charges       0.173532             0.014270   
Multiple Lines                          0.448428            -0.101236   
Internet Service                        0.287064             0.184681   
Avg Monthly GB Download                 0.063804             0.057459   
Online Security                         0.377673             0.148439   
Online Backup                           0.468395             0.076588   
Device Protection Plan                  0.461121             0.069854   
Premium Tech Support                    0.389047             0.151191   
Streaming TV                            0.213616            -0.129448   
Streaming Movies                        0.220091            -0.134447   
Streaming Music                         0.380614             0.041838   
Unlimited Data                         -0.057592            -0.034594   
Paperless Billing                       0.116670            -0.015241   
Monthly Charge                          0.554815            -0.058105   
Total Charges                           0.970266            -0.032446   
Total Refunds                           0.022574            -0.012800   
Total Extra Data Charges                0.114636             0.011158   
Total Long Distance Charges             0.765651            -0.070568   
Total Revenue                           1.000000            -0.046718   
Internet Type_Cable                    -0.046718             1.000000   
Internet Type_DSL                      -0.054856            -0.201234   
Internet Type_Fiber Optic               0.080137            -0.496382   
Contract_Month-to-Month                -0.451532            -0.006728   
Contract_One Year                       0.155290             0.010886   
Contract_Two Year                       0.355781            -0.002732   
Payment Method_Bank Withdrawal          0.062803            -0.035113   
Payment Method_Credit Card             -0.001403             0.012819   
Payment Method_Mailed Check            -0.137061             0.050307   

                                   Internet Type_DSL  \
Gender                                      0.014655   
Age                                        -0.056854   
Married                                     0.003602   
Number of Dependents                        0.010465   
Number of Referrals                         0.027511   
Tenure in Months                            0.016244   
Phone Service                              -0.334472   
Avg Monthly Long Distance Charges          -0.008801   
Multiple Lines                             -0.149177   
Internet Service                            0.279211   
Avg Monthly GB Download                     0.074321   
Online Security                             0.249760   
Online Backup                               0.121076   
Device Protection Plan                      0.120539   
Premium Tech Support                        0.236722   
Streaming TV                               -0.219388   
Streaming Movies                           -0.208604   
Streaming Music                             0.060038   
Unlimited Data                             -0.057505   
Paperless Billing                          -0.058684   
Monthly Charge                             -0.129892   
Total Charges                              -0.036622   
Total Refunds                               0.006062   
Total Extra Data Charges                    0.057450   
Total Long Distance Charges                -0.087915   
Total Revenue                              -0.054856   
Internet Type_Cable                        -0.201234   
Internet Type_DSL                           1.000000   
Internet Type_Fiber Optic                  -0.750457   
Contract_Month-to-Month                    -0.047333   
Contract_One Year                           0.040011   
Contract_Two Year                           0.015057   
Payment Method_Bank Withdrawal             -0.056924   
Payment Method_Credit Card                  0.053642   
Payment Method_Mailed Check                 0.009676   

                                   Internet Type_Fiber Optic  \
Gender                                             -0.007227   
Age                                                 0.078216   
Married                                            -0.007797   
Number of Dependents                               -0.013380   
Number of Referrals                                -0.030775   
Tenure in Months                                   -0.010868   
Phone Service                                       0.447803   
Avg Monthly Long Distance Charges                  -0.001829   
Multiple Lines                                      0.200512   
Internet Service                                   -0.372054   
Avg Monthly GB Download                            -0.104635   
Online Security                                    -0.321500   
Online Backup                                      -0.158977   
Device Protection Plan                             -0.153957   
Premium Tech Support                               -0.311804   
Streaming TV                                        0.281770   
Streaming Movies                                    0.275586   
Streaming Music                                    -0.081437   
Unlimited Data                                      0.074305   
Paperless Billing                                   0.062292   
Monthly Charge                                      0.154319   
Total Charges                                       0.054347   
Total Refunds                                       0.003264   
Total Extra Data Charges                           -0.058443   
Total Long Distance Charges                         0.125527   
Total Revenue                                       0.080137   
Internet Type_Cable                                -0.496382   
Internet Type_DSL                                  -0.750457   
Internet Type_Fiber Optic                           1.000000   
Contract_Month-to-Month                             0.046488   
Contract_One Year                                  -0.042804   
Contract_Two Year                                  -0.011501   
Payment Method_Bank Withdrawal                      0.074140   
Payment Method_Credit Card                         -0.056188   
Payment Method_Mailed Check                        -0.042518   

                                   Contract_Month-to-Month  Contract_One Year  \
Gender                                           -0.005772           0.008596   
Age                                              -0.001283          -0.010480   
Married                                          -0.252296           0.069478   
Number of Dependents                             -0.128389           0.004264   
Number of Referrals                              -0.272251           0.066514   
Tenure in Months                                 -0.614856           0.168878   
Phone Service                                    -0.002087           0.000181   
Avg Monthly Long Distance Charges                -0.002238           0.030331   
Multiple Lines                                   -0.070235          -0.020749   
Internet Service                                  0.223913          -0.035474   
Avg Monthly GB Download                           0.000653           0.019519   
Online Security                                  -0.225947           0.091162   
Online Backup                                    -0.154417           0.074896   
Device Protection Plan                           -0.203021           0.082671   
Premium Tech Support                             -0.256663           0.078670   
Streaming TV                                     -0.276596           0.079672   
Streaming Movies                                 -0.286515           0.082180   
Streaming Music                                  -0.067183           0.032663   
Unlimited Data                                   -0.050649           0.000779   
Paperless Billing                                 0.171170          -0.051747   
Monthly Charge                                    0.067649          -0.003094   
Total Charges                                    -0.418397           0.145967   
Total Refunds                                    -0.037443           0.004841   
Total Extra Data Charges                         -0.003023           0.021882   
Total Long Distance Charges                      -0.399747           0.131272   
Total Revenue                                    -0.451532           0.155290   
Internet Type_Cable                              -0.006728           0.010886   
Internet Type_DSL                                -0.047333           0.040011   
Internet Type_Fiber Optic                         0.046488          -0.042804   
Contract_Month-to-Month                           1.000000          -0.533797   
Contract_One Year                                -0.533797           1.000000   
Contract_Two Year                                -0.610011          -0.344435   
Payment Method_Bank Withdrawal                    0.163780          -0.053275   
Payment Method_Credit Card                       -0.186632           0.054667   
Payment Method_Mailed Check                       0.042806          -0.000710   

                                   Contract_Two Year  \
Gender                                     -0.001647   
Age                                         0.011245   
Married                                     0.214997   
Number of Dependents                        0.138544   
Number of Referrals                         0.239930   
Tenure in Months                            0.524372   
Phone Service                               0.002147   
Avg Monthly Long Distance Charges          -0.025937   
Multiple Lines                              0.097419   
Internet Service                           -0.215349   
Avg Monthly GB Download                    -0.019016   
Online Security                             0.165425   
Online Backup                               0.101252   
Device Protection Plan                      0.147928   
Premium Tech Support                        0.211232   
Streaming TV                                0.232423   
Streaming Movies                            0.241085   
Streaming Music                             0.043980   
Unlimited Data                              0.055501   
Paperless Billing                          -0.141546   
Monthly Charge                             -0.072206   
Total Charges                               0.327729   
Total Refunds                               0.037034   
Total Extra Data Charges                   -0.017149   
Total Long Distance Charges                 0.320794   
Total Revenue                               0.355781   
Internet Type_Cable                        -0.002732   
Internet Type_DSL                           0.015057   
Internet Type_Fiber Optic                  -0.011501   
Contract_Month-to-Month                    -0.610011   
Contract_One Year                          -0.344435   
Contract_Two Year                           1.000000   
Payment Method_Bank Withdrawal             -0.131909   
Payment Method_Credit Card                  0.155975   
Payment Method_Mailed Check                -0.046859   

                                   Payment Method_Bank Withdrawal  \
Gender                                                  -0.005451   
Age                                                      0.114948   
Married                                                 -0.003856   
Number of Dependents                                    -0.089444   
Number of Referrals                                     -0.053171   
Tenure in Months                                        -0.025183   
Phone Service                                            0.009661   
Avg Monthly Long Distance Charges                        0.013207   
Multiple Lines                                           0.132185   
Internet Service                                         0.259444   
Avg Monthly GB Download                                 -0.005962   
Online Security                                         -0.039649   
Online Backup                                            0.059466   
Device Protection Plan                                   0.055254   
Premium Tech Support                                    -0.035901   
Streaming TV                                            -0.049095   
Streaming Movies                                        -0.050527   
Streaming Music                                          0.121618   
Unlimited Data                                          -0.058125   
Paperless Billing                                        0.179961   
Monthly Charge                                           0.268894   
Total Charges                                            0.082373   
Total Refunds                                            0.026451   
Total Extra Data Charges                                 0.046901   
Total Long Distance Charges                             -0.009761   
Total Revenue                                            0.062803   
Internet Type_Cable                                     -0.035113   
Internet Type_DSL                                       -0.056924   
Internet Type_Fiber Optic                                0.074140   
Contract_Month-to-Month                                  0.163780   
Contract_One Year                                       -0.053275   
Contract_Two Year                                       -0.131909   
Payment Method_Bank Withdrawal                           1.000000   
Payment Method_Credit Card                              -0.897752   
Payment Method_Mailed Check                             -0.267501   

                                   Payment Method_Credit Card  \
Gender                                               0.007691   
Age                                                 -0.101897   
Married                                              0.033121   
Number of Dependents                                 0.099719   
Number of Referrals                                  0.077497   
Tenure in Months                                     0.084859   
Phone Service                                        0.004319   
Avg Monthly Long Distance Charges                   -0.010261   
Multiple Lines                                      -0.093775   
Internet Service                                    -0.225085   
Avg Monthly GB Download                              0.007092   
Online Security                                      0.051155   
Online Backup                                       -0.023816   
Device Protection Plan                              -0.028157   
Premium Tech Support                                 0.052222   
Streaming TV                                         0.057325   
Streaming Movies                                     0.062196   
Streaming Music                                     -0.092474   
Unlimited Data                                       0.052610   
Paperless Billing                                   -0.156198   
Monthly Charge                                      -0.215994   
Total Charges                                       -0.021773   
Total Refunds                                       -0.015421   
Total Extra Data Charges                            -0.033446   
Total Long Distance Charges                          0.053947   
Total Revenue                                       -0.001403   
Internet Type_Cable                                  0.012819   
Internet Type_DSL                                    0.053642   
Internet Type_Fiber Optic                           -0.056188   
Contract_Month-to-Month                             -0.186632   
Contract_One Year                                    0.054667   
Contract_Two Year                                    0.155975   
Payment Method_Bank Withdrawal                      -0.897752   
Payment Method_Credit Card                           1.000000   
Payment Method_Mailed Check                         -0.184299   

                                   Payment Method_Mailed Check  
Gender                                               -0.004661  
Age                                                  -0.033586  
Married                                              -0.063845  
Number of Dependents                                 -0.018555  
Number of Referrals                                  -0.050879  
Tenure in Months                                     -0.129430  
Phone Service                                        -0.031003  
Avg Monthly Long Distance Charges                    -0.007024  
Multiple Lines                                       -0.089814  
Internet Service                                     -0.086532  
Avg Monthly GB Download                              -0.002210  
Online Security                                      -0.023430  
Online Backup                                        -0.080588  
Device Protection Plan                               -0.061693  
Premium Tech Support                                 -0.034127  
Streaming TV                                         -0.015851  
Streaming Movies                                     -0.023309  
Streaming Music                                      -0.069081  
Unlimited Data                                        0.014612  
Paperless Billing                                    -0.059869  
Monthly Charge                                       -0.127502  
Total Charges                                        -0.136169  
Total Refunds                                        -0.025287  
Total Extra Data Charges                             -0.031488  
Total Long Distance Charges                          -0.096225  
Total Revenue                                        -0.137061  
Internet Type_Cable                                   0.050307  
Internet Type_DSL                                     0.009676  
Internet Type_Fiber Optic                            -0.042518  
Contract_Month-to-Month                               0.042806  
Contract_One Year                                    -0.000710  
Contract_Two Year                                    -0.046859  
Payment Method_Bank Withdrawal                       -0.267501  
Payment Method_Credit Card                           -0.184299  
Payment Method_Mailed Check                           1.000000  

[35 rows x 35 columns]


# Correlation Plot
plt.figure(figsize=(18, 18))
sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt=".2f", vmin=-1, vmax=1, annot_kws={"size": 10})
plt.xticks(rotation=90)
plt.yticks(rotation=0)
# Add title and adjust layout
plt.title('Correlation Matrix of Numerical Features')
plt.tight_layout()
# Show plot
plt.savefig('Correlation Matrix of Numerical Features.jpg')
plt.show()


# Checking for highly Correlated Features
# Setting a threshold for high correlation
threshold = 0.7
# Initializing a list to store highly correlated features pairs
highly_correlated_pairs = []
# Iterating through the correlation matrix
for i in range(len(correlation_matrix.columns)):
    for j in range(i+1, len(correlation_matrix.columns)):
        if abs(correlation_matrix.iloc[i, j]) > threshold:
            # Store the names of highly correlated features
            pair = (correlation_matrix.columns[i], correlation_matrix.columns[j])
            highly_correlated_pairs.append(pair)
if not highly_correlated_pairs:
    print("No highly correlated features found.")
else:
    print("Highly correlated feature pairs:")
    for pair in highly_correlated_pairs:
        print(pair)

Highly correlated feature pairs:
('Tenure in Months', 'Total Charges')
('Tenure in Months', 'Total Revenue')
('Internet Service', 'Monthly Charge')
('Unlimited Data', 'Total Extra Data Charges')
('Total Charges', 'Total Revenue')
('Total Long Distance Charges', 'Total Revenue')
('Internet Type_DSL', 'Internet Type_Fiber Optic')
('Payment Method_Bank Withdrawal', 'Payment Method_Credit Card')


# Removal of highly correlated features
Churn.drop(['Tenure in Months', 'Total Charges','Total Revenue', 'Internet Service', 'Monthly Charge','Unlimited Data', 
            'Total Extra Data Charges', 'Total Long Distance Charges', 'Internet Type_DSL', 'Internet Type_Fiber Optic', 'Payment Method_Bank Withdrawal', 'Payment Method_Credit Card'], axis =1, inplace= True)


#Checking Columns in dataset after feature removals
print(Churn.columns)

Index(['Gender', 'Age', 'Married', 'Number of Dependents',
       'Number of Referrals', 'Phone Service',
       'Avg Monthly Long Distance Charges', 'Multiple Lines',
       'Avg Monthly GB Download', 'Online Security', 'Online Backup',
       'Device Protection Plan', 'Premium Tech Support', 'Streaming TV',
       'Streaming Movies', 'Streaming Music', 'Paperless Billing',
       'Total Refunds', 'Customer Status', 'Internet Type_Cable',
       'Contract_Month-to-Month', 'Contract_One Year', 'Contract_Two Year',
       'Payment Method_Mailed Check'],
      dtype='object')


# Selecting only numerical columns
numerical_columns = Churn.select_dtypes(include=['int64', 'float64']).columns

# Calculate VIF
vif_data = pd.DataFrame()
vif_data["Feature"] = numerical_columns
vif_data["VIF"] = [variance_inflation_factor(Churn[numerical_columns].values, i) for i in range(len(numerical_columns))]

print("Variance Inflation Factors:")
print(vif_data)

Variance Inflation Factors:
                             Feature       VIF
0                                Age  2.338980
1               Number of Dependents  1.413638
2                Number of Referrals  1.535375
3  Avg Monthly Long Distance Charges  3.295581
4            Avg Monthly GB Download  2.371388
5                      Total Refunds  1.062786


# Iterate over each feature (excluding the target variable) in the dataset
for feature in Churn.columns[:-1]:
    # Check if the feature is integer-encoded (assuming categorical)
    if Churn[feature].dtype == 'int32':
        # Create a contingency table for the feature and the target variable
        contingency_table = pd.crosstab(Churn[feature], Churn['Customer Status'])
        
        # Perform the chi-square test
        chi2_statistic, p_value, _, _ = chi2_contingency(contingency_table)
        
        # Print the results
        print(f"Chi-square test results for {feature}:")
        print(f"   Chi-square statistic: {chi2_statistic}")
        print(f"   P-value: {p_value}")
        
        # Check for significance (e.g., p-value < 0.05)
        if p_value < 0.05:
            print(f"   {feature} is statistically significant for predicting the target variable.")
        else:
            print(f"   {feature} is not statistically significant for predicting the target variable.")
        
        print()

Chi-square test results for Gender:
   Chi-square statistic: 0.24007105217966312
   P-value: 0.6241548018547587
   Gender is not statistically significant for predicting the target variable.

Chi-square test results for Married:
   Chi-square statistic: 220.5060439643492
   P-value: 7.014602453362894e-50
   Married is statistically significant for predicting the target variable.

Chi-square test results for Phone Service:
   Chi-square statistic: 1.2551734871989866
   P-value: 0.26256666476273127
   Phone Service is not statistically significant for predicting the target variable.

Chi-square test results for Multiple Lines:
   Chi-square statistic: 1.1535971852605198
   P-value: 0.2827973969037025
   Multiple Lines is not statistically significant for predicting the target variable.

Chi-square test results for Online Security:
   Chi-square statistic: 248.43749367640638
   P-value: 5.689634112881904e-56
   Online Security is statistically significant for predicting the target variable.

Chi-square test results for Online Backup:
   Chi-square statistic: 73.07537713497942
   P-value: 1.247916969909602e-17
   Online Backup is statistically significant for predicting the target variable.

Chi-square test results for Device Protection Plan:
   Chi-square statistic: 56.671577391479005
   P-value: 5.1502608160860495e-14
   Device Protection Plan is statistically significant for predicting the target variable.

Chi-square test results for Premium Tech Support:
   Chi-square statistic: 231.66761732586693
   P-value: 2.580290732051722e-52
   Premium Tech Support is statistically significant for predicting the target variable.

Chi-square test results for Streaming TV:
   Chi-square statistic: 135.09725737576034
   P-value: 3.1438574036725362e-31
   Streaming TV is statistically significant for predicting the target variable.

Chi-square test results for Streaming Movies:
   Chi-square statistic: 139.63452425587423
   P-value: 3.199893781319627e-32
   Streaming Movies is statistically significant for predicting the target variable.

Chi-square test results for Streaming Music:
   Chi-square statistic: 4.274339099232788
   P-value: 0.03869200784160805
   Streaming Music is statistically significant for predicting the target variable.

Chi-square test results for Paperless Billing:
   Chi-square statistic: 231.29426752260608
   P-value: 3.112345421939546e-52
   Paperless Billing is statistically significant for predicting the target variable.

Chi-square test results for Customer Status:
   Chi-square statistic: 6584.07952972205
   P-value: 0.0
   Customer Status is statistically significant for predicting the target variable.

Chi-square test results for Internet Type_Cable:
   Chi-square statistic: 0.26359201971038076
   P-value: 0.6076630592581913
   Internet Type_Cable is not statistically significant for predicting the target variable.

Chi-square test results for Contract_Month-to-Month:
   Chi-square statistic: 1665.0430834588672
   P-value: 0.0
   Contract_Month-to-Month is statistically significant for predicting the target variable.

Chi-square test results for Contract_One Year:
   Chi-square statistic: 297.7642870461098
   P-value: 1.0112624954818955e-66
   Contract_One Year is statistically significant for predicting the target variable.

Chi-square test results for Contract_Two Year:
   Chi-square statistic: 846.9257076455871
   P-value: 3.3878596704718002e-186
   Contract_Two Year is statistically significant for predicting the target variable.


# Removal of non statistically significant features
Churn.drop(["Gender", "Phone Service", "Multiple Lines", "Internet Type_Cable"], axis =1, inplace= True)


# Checking Shape of dataset after removal of irrelevant features
Churn.shape

(6589, 20)


#Splitting the dataset into features (X) and target variable (Y)
X = Churn.drop(columns=['Customer Status'], axis=1)
Y = Churn["Customer Status"]


# Shuffle your data
X,Y = shuffle(X, Y, random_state=42)


X.head()


Y.head()

2059    0
4323    1
1544    1
2644    1
4029    1
Name: Customer Status, dtype: int32


import statsmodels.api as sm
logit_model=sm.Logit(Y,X)
result=logit_model.fit()
print(result.summary2())

Optimization terminated successfully.
         Current function value: 0.365385
         Iterations 8
                                  Results: Logit
==================================================================================
Model:                    Logit                  Method:                 MLE      
Dependent Variable:       Customer Status        Pseudo R-squared:       0.387    
Date:                     2024-05-05 21:37       AIC:                    4853.0382
No. Observations:         6589                   BIC:                    4982.1082
Df Model:                 18                     Log-Likelihood:         -2407.5  
Df Residuals:             6570                   LL-Null:                -3929.5  
Converged:                1.0000                 LLR p-value:            0.0000   
No. Iterations:           8.0000                 Scale:                  1.0000   
----------------------------------------------------------------------------------
                                   Coef.  Std.Err.    z     P>|z|   [0.025  0.975]
----------------------------------------------------------------------------------
Age                               -1.4799   0.1598  -9.2624 0.0000 -1.7931 -1.1668
Married                           -1.4031   0.1056 -13.2850 0.0000 -1.6101 -1.1961
Number of Dependents               6.2976   0.5564  11.3175 0.0000  5.2070  7.3882
Number of Referrals                5.9307   0.3722  15.9355 0.0000  5.2013  6.6602
Avg Monthly Long Distance Charges  0.0518   0.1300   0.3983 0.6904 -0.2031  0.3067
Avg Monthly GB Download           -0.0047   0.0025  -1.9219 0.0546 -0.0095  0.0001
Online Security                    0.5956   0.0909   6.5535 0.0000  0.4174  0.7737
Online Backup                      0.2666   0.0814   3.2754 0.0011  0.1071  0.4261
Device Protection Plan            -0.0436   0.0838  -0.5205 0.6027 -0.2079  0.1206
Premium Tech Support               0.4536   0.0911   4.9818 0.0000  0.2752  0.6321
Streaming TV                      -0.1720   0.0849  -2.0261 0.0428 -0.3383 -0.0056
Streaming Movies                   0.7710   0.1100   7.0106 0.0000  0.5554  0.9865
Streaming Music                   -1.0547   0.1081  -9.7557 0.0000 -1.2666 -0.8428
Paperless Billing                 -0.4579   0.0792  -5.7805 0.0000 -0.6131 -0.3026
Total Refunds                      0.0095   0.0048   1.9988 0.0456  0.0002  0.0189
Contract_Month-to-Month            0.6522   0.1538   4.2414 0.0000  0.3508  0.9536
Contract_One Year                  2.6115   0.1885  13.8548 0.0000  2.2420  2.9809
Contract_Two Year                  3.8988   0.2299  16.9601 0.0000  3.4482  4.3493
Payment Method_Mailed Check       -0.7775   0.1495  -5.2020 0.0000 -1.0705 -0.4846
==================================================================================


#Removing features with P values higher than 0.05
# List of columns to drop
columns_to_drop = ['Avg Monthly Long Distance Charges','Avg Monthly GB Download', 'Device Protection Plan']
# Drop the specified columns from the DataFrame X
X = X.drop(columns_to_drop, axis=1)


logit_model=sm.Logit(Y,X)
result2=logit_model.fit()
print(result2.summary2())

Optimization terminated successfully.
         Current function value: 0.365699
         Iterations 8
                               Results: Logit
============================================================================
Model:                  Logit                Method:               MLE      
Dependent Variable:     Customer Status      Pseudo R-squared:     0.387    
Date:                   2024-05-05 21:37     AIC:                  4851.1792
No. Observations:       6589                 BIC:                  4959.8697
Df Model:               15                   Log-Likelihood:       -2409.6  
Df Residuals:           6573                 LL-Null:              -3929.5  
Converged:              1.0000               LLR p-value:          0.0000   
No. Iterations:         8.0000               Scale:                1.0000   
----------------------------------------------------------------------------
                             Coef.  Std.Err.    z     P>|z|   [0.025  0.975]
----------------------------------------------------------------------------
Age                         -1.3355   0.1384  -9.6518 0.0000 -1.6067 -1.0643
Married                     -1.4060   0.1055 -13.3273 0.0000 -1.6128 -1.1992
Number of Dependents         6.1496   0.5511  11.1589 0.0000  5.0695  7.2298
Number of Referrals          5.9200   0.3718  15.9218 0.0000  5.1913  6.6488
Online Security              0.5932   0.0908   6.5345 0.0000  0.4153  0.7711
Online Backup                0.2601   0.0810   3.2106 0.0013  0.1013  0.4189
Premium Tech Support         0.4500   0.0905   4.9719 0.0000  0.2726  0.6274
Streaming TV                -0.1733   0.0846  -2.0491 0.0405 -0.3390 -0.0075
Streaming Movies             0.7908   0.1091   7.2481 0.0000  0.5769  1.0046
Streaming Music             -1.0849   0.1056 -10.2775 0.0000 -1.2918 -0.8780
Paperless Billing           -0.4611   0.0791  -5.8289 0.0000 -0.6162 -0.3061
Total Refunds                0.0095   0.0048   1.9952 0.0460  0.0002  0.0189
Contract_Month-to-Month      0.4972   0.1044   4.7645 0.0000  0.2927  0.7018
Contract_One Year            2.4482   0.1481  16.5328 0.0000  2.1580  2.7384
Contract_Two Year            3.7308   0.1980  18.8387 0.0000  3.3426  4.1189
Payment Method_Mailed Check -0.7758   0.1493  -5.1955 0.0000 -1.0685 -0.4832
============================================================================


# Splitting the data into training and testing sets
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.3, random_state=42)

# Checking the shape of the resulting sets
print("Shape of X_train:", X_train.shape)
print("Shape of X_test:", X_test.shape)
print("Shape of Y_train:", Y_train.shape)
print("Shape of Y_test:", Y_test.shape)

Shape of X_train: (4612, 16)
Shape of X_test: (1977, 16)
Shape of Y_train: (4612,)
Shape of Y_test: (1977,)


# Count the occurrences of each class
class_counts = Y_train.value_counts()

# Reorder the class counts so that Churned (0) is displayed before Stayed (1)
class_counts_reordered = class_counts.reindex([0, 1])

# Plot the class distribution
plt.figure(figsize=(8, 6))
class_counts_reordered.plot(kind='bar', color=['red', 'skyblue',])
plt.xlabel('Customer Status')
plt.ylabel('Count')
plt.title('Class Distribution before Upsampling')
plt.xticks([0, 1], ['Churned', 'Stayed'], rotation=0)  # Set x-axis ticks to display class labels
plt.savefig("Target variable distribution.jpg")
plt.show()


# Initialize SMOTE with a desired sampling strategy
smote = SMOTE(sampling_strategy='auto', random_state=42)
# Upsample the minority class in the training data
X_train, Y_train = smote.fit_resample(X_train, Y_train)
X_test, Y_test = smote.fit_resample(X_test, Y_test)


# Count the occurrences of each class
class_counts = Y_train.value_counts()
# Reorder the class counts so that Churned (0) is displayed before Stayed (1)
class_counts_reordered = class_counts.reindex([0, 1])
# Plot the class distribution
plt.figure(figsize=(8, 6))
class_counts_reordered.plot(kind='bar', color=['red', 'skyblue',])
plt.xlabel('Customer Status')
plt.ylabel('Count')
plt.title('Class Distribution after Upsampling')
plt.xticks([0, 1], ['Churned', 'Stayed'], rotation=0) 
plt.savefig("Distribution after SMOTE.jpg")
plt.show()


Y_train.value_counts()

Customer Status
1    3327
0    3327
Name: count, dtype: int64


Y_test.value_counts()

Customer Status
0    1393
1    1393
Name: count, dtype: int64


# Train a Logistic Regression Model (using the default parameters)
log_model = LogisticRegression(random_state=16, max_iter=1000)
# fit the model with data
log_model.fit(X_train, Y_train)

LogisticRegression(max_iter=1000, random_state=16)

LogisticRegression(max_iter=1000, random_state=16)


y_pred_log = log_model.predict(X_test)


#printing accuracy
accuracy_log = accuracy_score(Y_test, y_pred_log)
precision_log = precision_score(Y_test, y_pred_log)
recall_log = recall_score(Y_test, y_pred_log)
f1_log = f1_score(Y_test, y_pred_log)
print("Accuracy:", accuracy_log)
print("Precision:", precision_log)
print("Recall:", recall_log)
print("F1 Score:", f1_log)
# Print classification report
print("Logistic Regression Classification Report before optimisation:")
print(classification_report(Y_test, y_pred_log))

Accuracy: 0.8201722900215362
Precision: 0.8534072900158478
Recall: 0.7731514716439339
F1 Score: 0.8112994350282486
Logistic Regression Classification Report before optimisation:
              precision    recall  f1-score   support

           0       0.79      0.87      0.83      1393
           1       0.85      0.77      0.81      1393

    accuracy                           0.82      2786
   macro avg       0.82      0.82      0.82      2786
weighted avg       0.82      0.82      0.82      2786


# Define the parameter grid
param_grid = {
    'C': [0.001, 0.01, 0.1, 1, 10, 100],  
    'penalty': ['l1', 'l2'],               
    'solver': ['liblinear', 'saga']        
}


warnings.filterwarnings("ignore", category=ConvergenceWarning)
# Creating a logistic regression model
logistic_regression = LogisticRegression(max_iter=1000, random_state=16)
# The GridSearchCV object
grid_search = GridSearchCV(logistic_regression, param_grid, cv=5, scoring='accuracy')
# Performing grid search to find the best hyperparameters
grid_search.fit(X_train, Y_train)

GridSearchCV(cv=5, estimator=LogisticRegression(max_iter=1000, random_state=16),
             param_grid={'C': [0.001, 0.01, 0.1, 1, 10, 100],
                         'penalty': ['l1', 'l2'],
                         'solver': ['liblinear', 'saga']},
             scoring='accuracy')

GridSearchCV(cv=5, estimator=LogisticRegression(max_iter=1000, random_state=16),
             param_grid={'C': [0.001, 0.01, 0.1, 1, 10, 100],
                         'penalty': ['l1', 'l2'],
                         'solver': ['liblinear', 'saga']},
             scoring='accuracy')

LogisticRegression(max_iter=1000, random_state=16)

LogisticRegression(max_iter=1000, random_state=16)


# Printing the best hyperparameters found
print("Best hyperparameters:", grid_search.best_params_)

Best hyperparameters: {'C': 100, 'penalty': 'l1', 'solver': 'liblinear'}


# Get the best model
best_model = grid_search.best_estimator_
# Evaluate the best model on the test set
y_pred_best = best_model.predict(X_test)


# Calculate accuracy
accuracy_logreg = accuracy_score(Y_test, y_pred_best)
precision_logreg = precision_score(Y_test, y_pred_best)
recall_logreg = recall_score(Y_test, y_pred_best)
f1_logreg = f1_score(Y_test, y_pred_best)
print("Accuracy of best model:", accuracy_logreg)
print("Precision of best model:", precision_logreg)
print("Recall of best model:", recall_logreg)
print("F1 of best model:", f1_logreg)
# Print classification report
print("Logistic Regression Classification Report:")
print(classification_report(Y_test, y_pred_best))

Accuracy of best model: 0.8230437903804738
Precision of best model: 0.8543307086614174
Recall of best model: 0.7788944723618091
F1 of best model: 0.8148704468644387
Logistic Regression Classification Report:
              precision    recall  f1-score   support

           0       0.80      0.87      0.83      1393
           1       0.85      0.78      0.81      1393

    accuracy                           0.82      2786
   macro avg       0.83      0.82      0.82      2786
weighted avg       0.83      0.82      0.82      2786


# Plot the confusion matrix for the Logistic Regression Model
cm_log = confusion_matrix(Y_test, y_pred_best)
disp_log = metrics.ConfusionMatrixDisplay(confusion_matrix=cm_log);
disp_log.plot(cmap='Oranges')  
plt.title('Confusion Matrix (Logistic Regression)')
plt.xlabel('Predicted Label')
plt.ylabel('True Label')
plt.savefig('Confusion Matrix for Logistic Regression.jpg')
plt.show()


# Calculate Precision and Recall for the optimized model
precision_best, recall_best, thresholds_best = precision_recall_curve(Y_test, y_pred_best)
# Create Precision-Recall Curve
fig, ax = plt.subplots()
ax.plot(recall_best, precision_best, color='purple')
# Adding axis labels and title to the plot
ax.set_title('Precision-Recall Curve for Optimised Logistic Regression Model')
ax.set_ylabel('Precision')
ax.set_xlabel('Recall')
# Save the plot as an image file and display
plt.savefig('Precision-Recall Curve for Optimised Logistic Regression Model.jpg')
plt.show()


# Calculate False Positive Rate, True Positive Rate, and Thresholds for the optimized model
fpr_best, tpr_best, thresholds_best = roc_curve(Y_test, y_pred_best)
# Calculate the AUC score for the optimized model
auc_score_best = roc_auc_score(Y_test, y_pred_best)
# Plot ROC curve for the optimized model
plt.figure(figsize=(8, 6))
plt.plot(fpr_best, tpr_best, color='blue', lw=2, label=f'AUC = {auc_score_best:.2f}')
plt.plot([0, 1], [0, 1], color='red', linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) Curve for Optimised Logistic Regression Model')
plt.legend(loc='center')
plt.grid(True)
plt.savefig('Receiver Operating Characteristic (ROC) Curve for Optimised Logistic Regression Model.jpg')
plt.show()


# Train a Random Forest model
rf_model = RandomForestClassifier()
rf_model.fit(X_train, Y_train)

RandomForestClassifier()

RandomForestClassifier()


y_pred_rf = rf_model.predict(X_test)


#printing accuracy
accuracy = accuracy_score(Y_test, y_pred_rf)
print("Accuracy:", accuracy)
# Print classification report
print("Random Forest Classification Report before optimisation:")
print(classification_report(Y_test, y_pred_rf))

Accuracy: 0.7954055994256999
Random Forest Classification Report before optimisation:
              precision    recall  f1-score   support

           0       0.83      0.74      0.78      1393
           1       0.76      0.85      0.81      1393

    accuracy                           0.80      2786
   macro avg       0.80      0.80      0.79      2786
weighted avg       0.80      0.80      0.79      2786


# Convert X_train to a pandas DataFrame
X_train_df = pd.DataFrame(X_train)

from sklearn.tree import plot_tree

# Export the first three decision trees from the forest
for i in range(3):
    plt.figure(figsize=(10, 5))
    tree = rf_model.estimators_[i]
    plot_tree(tree,
              feature_names=X_train_df.columns, 
              filled=True,
              max_depth=2,
              impurity=False,
              proportion=True)
    plt.show()


# Define the parameter grid for RandomForestClassifier
param_grid = {
    'n_estimators': [100, 200, 300],
    'max_depth': [None, 10, 20],
    'min_samples_split': [2, 5, 10],
    'min_samples_leaf': [1, 2, 4]
}
# The grid search
grid_search = GridSearchCV(estimator=RandomForestClassifier(), param_grid=param_grid, cv=3)
# Fit the grid search to the data
grid_search.fit(X_train, Y_train)

GridSearchCV(cv=3, estimator=RandomForestClassifier(),
             param_grid={'max_depth': [None, 10, 20],
                         'min_samples_leaf': [1, 2, 4],
                         'min_samples_split': [2, 5, 10],
                         'n_estimators': [100, 200, 300]})

GridSearchCV(cv=3, estimator=RandomForestClassifier(),
             param_grid={'max_depth': [None, 10, 20],
                         'min_samples_leaf': [1, 2, 4],
                         'min_samples_split': [2, 5, 10],
                         'n_estimators': [100, 200, 300]})

RandomForestClassifier()

RandomForestClassifier()


# Get the best parameters and the best estimator
best_params = grid_search.best_params_
best_estimator = grid_search.best_estimator_
# Print the best parameters
print("Best Parameters:", best_params)

Best Parameters: {'max_depth': 20, 'min_samples_leaf': 2, 'min_samples_split': 5, 'n_estimators': 100}


# Generate predictions with the best model
y_pred_best_rf = best_estimator.predict(X_test)
accuracy_rf = accuracy_score(Y_test, y_pred_best_rf)
precision_rf = precision_score(Y_test, y_pred_best_rf)
recall_rf = recall_score(Y_test, y_pred_best_rf)
f1_rf = f1_score(Y_test, y_pred_best_rf)

print("Accuracy:", accuracy_rf)
print("Precision:", precision_rf)
print("Recall:", recall_rf)
print("F1 Score:", f1_rf)
# Print classification report
print("Random Forest Classification Report:")
print(classification_report(Y_test, y_pred_best_rf))

Accuracy: 0.8277099784637473
Precision: 0.8217054263565892
Recall: 0.8370423546302943
F1 Score: 0.829302987197724
Random Forest Classification Report:
              precision    recall  f1-score   support

           0       0.83      0.82      0.83      1393
           1       0.82      0.84      0.83      1393

    accuracy                           0.83      2786
   macro avg       0.83      0.83      0.83      2786
weighted avg       0.83      0.83      0.83      2786


# Plot the confusion matrix for the Random Forest Model
cm = confusion_matrix(Y_test, y_pred_best_rf)
disp_rf = metrics.ConfusionMatrixDisplay(confusion_matrix=cm);
disp_rf.plot(cmap='Greens')  
plt.title('Confusion Matrix (Random Forest)')
plt.xlabel('Predicted Label')
plt.ylabel('True Label')
plt.savefig("Confusion Matrix for Random Forest")
plt.show()


# Calculate Precision and Recall
precision, recall, thresholds = precision_recall_curve(Y_test, best_estimator.predict_proba(X_test)[:, 1])
# Create Precision Recall Curve
fig, ax = plt.subplots()
ax.plot(recall, precision, color='purple')
# Adding axis labels and title to the plot
ax.set_title('Precision-Recall Curve for Random Forest')
ax.set_ylabel('Precision')
ax.set_xlabel('Recall')
plt.savefig('Precision-Recall Curve for Random Forest.jpg')
plt.show()


# Calculate the false positive rate, true positive rate, and thresholds
fpr, tpr, thresholds = roc_curve(Y_test, y_pred_best_rf)
# Calculate the AUC score
auc_score = roc_auc_score(Y_test, y_pred_best_rf)
# Plot ROC curve
plt.figure(figsize=(8, 6))
plt.plot(fpr, tpr, color='blue', lw=2, label=f'AUC = {auc_score:.2f}')
plt.plot([0, 1], [0, 1], color='red', linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) Curve for Random Forest Classifier')
plt.legend(loc='center')
plt.grid(True)
plt.savefig('Receiver Operating Characteristic (ROC) Curve for Random Forest Classifier.jpg')
plt.show()


# Initializing the KNN classifier with a specified number of neighbors (e.g., 3)
knn = KNeighborsClassifier(n_neighbors=3)


# Training the classifier on the training data
knn.fit(X_train, Y_train)

KNeighborsClassifier(n_neighbors=3)

KNeighborsClassifier(n_neighbors=3)


#Evaluating the performance of the model using the test set
knn_Y_predict = knn.predict(X_test)


# Model Accuracy: how often is the classifier correct?
print("Accuracy:",metrics.accuracy_score(Y_test, knn_Y_predict))

Accuracy: 0.7613065326633166


# Print classification report
print("KNN Classification Report:")
print(classification_report(Y_test,knn_Y_predict))

KNN Classification Report:
              precision    recall  f1-score   support

           0       0.77      0.74      0.76      1393
           1       0.75      0.78      0.77      1393

    accuracy                           0.76      2786
   macro avg       0.76      0.76      0.76      2786
weighted avg       0.76      0.76      0.76      2786


# Defining the range of values for n_neighbors
neighbors = np.arange(1, 36)


# Create an empty list to store the accuracy at different values of K
KNN_accuracy = []


# Loop through each value of n_neighbors and perform grid search
for k in neighbors:
    knn = KNeighborsClassifier(n_neighbors=k)
    grid_search = GridSearchCV(knn, {'n_neighbors': [k]}, cv=5)
    grid_search.fit(X_train, Y_train)
    KNN_accuracy.append(grid_search.best_score_)


# Plot the mean test scores
plt.figure(figsize=(10, 6))
plt.plot(neighbors, KNN_accuracy, marker='o')
plt.title('KNN Accuracy for Different Values of K')
plt.xlabel('Number of Neighbors (K)')
plt.ylabel('Accuracy')
plt.grid(True)
plt.xticks(np.arange(0, 36, step=2))
plt.savefig('KNN Accuracy for Different Values of K.jpg')
plt.show()


best_index = np.argmax(KNN_accuracy)
best_k = neighbors[best_index]

print(f"The best value of n_neighbors (K) is: {best_k}")

The best value of n_neighbors (K) is: 3


knn= KNeighborsClassifier(n_neighbors=best_k)
knn.fit(X_train,Y_train)
best_k_predictions = knn.predict(X_test)


# Calculate precision and recall for the best K value
accuracy_knn = accuracy_score(Y_test, best_k_predictions)
precision_knn = precision_score(Y_test, best_k_predictions)
recall_knn = recall_score(Y_test, best_k_predictions)
f1_knn = f1_score(Y_test, best_k_predictions)
# Print the accuracy, precision, and recall for the best K value
print(f"Best K value: {best_k}")
print(f"Accuracy for best K value: {accuracy_knn}")
print(f"Precision for best K value: {precision_knn}")
print(f"Recall for best K value: {recall_knn}")
print(f"F1 Score for best K value: {f1_knn}")

# Print classification report
print("KNN Classification Report:")
print(classification_report(Y_test ,best_k_predictions))

Best K value: 3
Accuracy for best K value: 0.7613065326633166
Precision for best K value: 0.7506887052341598
Recall for best K value: 0.782483847810481
F1 Score for best K value: 0.7662565905096661
KNN Classification Report:
              precision    recall  f1-score   support

           0       0.77      0.74      0.76      1393
           1       0.75      0.78      0.77      1393

    accuracy                           0.76      2786
   macro avg       0.76      0.76      0.76      2786
weighted avg       0.76      0.76      0.76      2786


# Plot the confusion matrix for the KNN Model
cm_knn = confusion_matrix(Y_test, best_k_predictions)
disp_knn = metrics.ConfusionMatrixDisplay(confusion_matrix=cm_knn);
disp_knn.plot(cmap='Purples')  
plt.title(f'Confusion Matrix for KNN (K={best_k})')
plt.xlabel('Predicted Label')
plt.ylabel('True Label')
plt.savefig('Confusion Matrix for KNN.jpg')
plt.show()


# Calculate Precision and Recall
precision, recall, thresholds = precision_recall_curve(Y_test, knn.predict_proba(X_test)[:, 1])
# Create Precision Recall Curve
fig, ax = plt.subplots()
ax.plot(recall, precision, color='purple')
# Adding axis labels and title to the plot
ax.set_title('Precision-Recall Curve for KNN')
ax.set_ylabel('Precision')
ax.set_xlabel('Recall')
plt.savefig('Precision-Recall Curve for KNN.jpg')
plt.show()


# Calculate the false positive rate, true positive rate, and thresholds
fpr, tpr, thresholds = roc_curve(Y_test, best_k_predictions)
# Calculate the AUC score
auc_score = roc_auc_score(Y_test, best_k_predictions)
# Plot ROC curve
plt.figure(figsize=(8, 6))
plt.plot(fpr, tpr, color='blue', lw=2, label=f'AUC = {auc_score:.2f}')
plt.plot([0, 1], [0, 1], color='red', linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) Curve for KNN')
plt.legend(loc='center')
plt.grid(True)
plt.savefig('Receiver Operating Characteristic (ROC) Curve for KNN.jpg')
plt.show()


# Initializing the GNB classifier
GNB = GaussianNB()


# Fitting the classifier to the training data
GNB.fit(X_train, Y_train)

GaussianNB()

GaussianNB()


# Predicting the target variable for the test data
gnb_Y_predict = GNB.predict(X_test)


# Calculate precision and recall for the GNB Model
accuracy_gnb = accuracy_score(Y_test, gnb_Y_predict)
precision_gnb = precision_score(Y_test, gnb_Y_predict)
recall_gnb= recall_score(Y_test, gnb_Y_predict)
f1_gnb= f1_score(Y_test, gnb_Y_predict)
# Print the accuracy, precision, and recall for GNB Model
print(f"Accuracy for GNB Model: {accuracy_gnb} ")
print(f"Precision for GNB Model: {precision_gnb}")
print(f"Recall for GNB Model: {recall_gnb}")
print(f"F1 Score for GNB Model: {f1_gnb}")

Accuracy for GNB Model: 0.7989949748743719 
Precision for GNB Model: 0.8078344419807835
Recall for GNB Model: 0.7846374730796841
F1 Score for GNB Model: 0.7960670065549891


# Print classification report
print("GNB Classification Report:")
print(classification_report(Y_test, gnb_Y_predict))

GNB Classification Report:
              precision    recall  f1-score   support

           0       0.79      0.81      0.80      1393
           1       0.81      0.78      0.80      1393

    accuracy                           0.80      2786
   macro avg       0.80      0.80      0.80      2786
weighted avg       0.80      0.80      0.80      2786


# Define the parameter grid
param_grid = {'var_smoothing': np.logspace(0,-9, num=100)}
# Initialize GridSearchCV
grid_search = GridSearchCV(GNB, param_grid, cv=5, scoring='accuracy')


# Fit the grid search to the data
grid_search.fit(X_train, Y_train)
# Get the best parameters
best_params = grid_search.best_params_


# Initialize GNB with the best parameters
optimised_GNB = GaussianNB(var_smoothing=best_params['var_smoothing'])
# Fit the optimized model to the training data
optimised_GNB.fit(X_train, Y_train)
# Predict the target variable for the test data
optimised_gnb_Y_predict = optimised_GNB.predict(X_test)


# Calculate and print accuracy, precision, and recall for the optimized GNB model
accuracy_gnb = accuracy_score(Y_test, optimised_gnb_Y_predict)
precision_gnb = precision_score(Y_test, optimised_gnb_Y_predict)
recall_gnb = recall_score(Y_test, optimised_gnb_Y_predict)
f1_gnb = f1_score(Y_test, optimised_gnb_Y_predict)

print(f"Accuracy for Optimised GNB Model: {accuracy_gnb}")
print(f"Precision for Optimised GNB Model: {precision_gnb}")
print(f"Recall for Optimised GNB Model: {recall_gnb}")
print(f"F1 Score for Optimised GNB Model: {f1_gnb}")
# Print classification report
print("GNB Classification Report:")
print(classification_report(Y_test,  optimised_gnb_Y_predict))

Accuracy for Optimised GNB Model: 0.7982770997846375
Precision for Optimised GNB Model: 0.8070953436807096
Recall for Optimised GNB Model: 0.7839195979899497
F1 Score for Optimised GNB Model: 0.7953386744355426
GNB Classification Report:
              precision    recall  f1-score   support

           0       0.79      0.81      0.80      1393
           1       0.81      0.78      0.80      1393

    accuracy                           0.80      2786
   macro avg       0.80      0.80      0.80      2786
weighted avg       0.80      0.80      0.80      2786


# Plot the confusion matrix for the GNB Model
cm_gnb = confusion_matrix(Y_test,  optimised_gnb_Y_predict)
disp_gnb = metrics.ConfusionMatrixDisplay(confusion_matrix=cm_knn);
disp_gnb.plot(cmap='Reds')  
plt.title(f'Confusion Matrix for GNB Model)')
plt.xlabel('Predicted Label')
plt.ylabel('True Label')
plt.savefig('Confusion Matrix for GNB.jpg')
plt.show()


# Calculate Precision and Recall
precision, recall, thresholds = precision_recall_curve(Y_test, optimised_GNB.predict_proba(X_test)[:, 1])
# Create Precision Recall Curve
fig, ax = plt.subplots()
ax.plot(recall, precision, color='purple')
# Adding axis labels and title to the plot
ax.set_title('Precision-Recall Curve for GNB')
ax.set_ylabel('Precision')
ax.set_xlabel('Recall')
plt.savefig('Precision-Recall Curve for GNB.jpg')
plt.show()


# Calculate the false positive rate, true positive rate, and thresholds
fpr, tpr, thresholds = roc_curve(Y_test, optimised_gnb_Y_predict)
# Calculate the AUC score
auc_score = roc_auc_score(Y_test, optimised_gnb_Y_predict)
# Plot ROC curve
plt.figure(figsize=(8, 6))
plt.plot(fpr, tpr, color='blue', lw=2, label=f'AUC = {auc_score:.2f}')
plt.plot([0, 1], [0, 1], color='red', linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) Curve for Guassian Naives Bayes')
plt.legend(loc='center')
plt.grid(True)
plt.savefig('Receiver Operating Characteristic (ROC) Curve for GNB.jpg')
plt.show()


# Classifier names
classifiers = ['Logistic_Regression', 'Random_Forest', 'KNN', 'Gaussian Naive Bayes']
# Accuracy scores obtained from cross-validation or testing
accuracy_scores = [accuracy_logreg, accuracy_rf, accuracy_knn, accuracy_gnb]
# Define colors for each classifier
colors = ['skyblue', 'green', 'red', 'purple']
# Create a bar plot
plt.figure(figsize=(10, 6))
bars = plt.barh(classifiers, accuracy_scores, color=colors)
plt.xlabel('Accuracy')
plt.title('Accuracy of Different Classifiers')
plt.gca().invert_yaxis()  
# Add labels on bars
for bar, score in zip(bars, accuracy_scores):
    plt.text(bar.get_width(), bar.get_y() + bar.get_height()/2, f'{score:.2f}', 
             va='center', ha='left', fontsize=9)
plt.savefig('Model Comparision.jpg')
plt.show()


# Calculate ROC curves and AUC scores for each model
models = {
    'Random Forest': (y_pred_best_rf, 'blue'),
    'Logistic Regression': (y_pred_best, 'orange'),
    'KNN': (best_k_predictions, 'green'),
    'Gaussian Naive Bayes': (optimised_gnb_Y_predict, 'purple')
}
plt.figure(figsize=(10, 8))
# Plot ROC curves for each model
for model_name, (y_pred, color) in models.items():
    fpr, tpr, _ = roc_curve(Y_test, y_pred)
    auc_score = roc_auc_score(Y_test, y_pred)
    plt.plot(fpr, tpr, lw=2, label=f'{model_name} (AUC = {auc_score:.2f})', color=color)
# Plot the diagonal line (no-skill line)
plt.plot([0, 1], [0, 1], linestyle='--', color='red', lw=2)
# Add legend, labels, and title
plt.legend(loc='lower right')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) Curve')
plt.savefig('Combined_ROC_Curve.png')
plt.show()


# Define the results dictionary
results = {
    'Model': ['Logistic Regression', 'Random Forest', 'KNN', 'Naive Bayes'],
    'Accuracy %': [accuracy_score(Y_test, y_pred_best), accuracy_score(Y_test, y_pred_best_rf), accuracy_score(Y_test, best_k_predictions), accuracy_score(Y_test, optimised_gnb_Y_predict)],
    'Precision %': [precision_score(Y_test, y_pred_best), precision_score(Y_test, y_pred_best_rf), precision_score(Y_test, best_k_predictions), precision_score(Y_test, optimised_gnb_Y_predict)],
    'Recall %': [recall_score(Y_test, y_pred_best), recall_score(Y_test, y_pred_best_rf), recall_score(Y_test, best_k_predictions), recall_score(Y_test, optimised_gnb_Y_predict)],
    'F1-Score': [f"{f1_score(Y_test, y_pred_best):.2f}", f"{f1_score(Y_test, y_pred_best_rf):.2f}", f"{f1_score(Y_test, best_k_predictions):.2f}", f"{f1_score(Y_test, optimised_gnb_Y_predict):.2f}"]
}
# Convert accuracy, precision, recall scores to percentages and F1 scores to strings with 2 decimal places
results['Accuracy %'] = [f"{score * 100:.2f}%" for score in results['Accuracy %']]
results['Precision %'] = [f"{score * 100:.2f}%" for score in results['Precision %']]
results['Recall %'] = [f"{score * 100:.2f}%" for score in results['Recall %']]
# Create a DataFrame
results_df = pd.DataFrame(results)
# Display the DataFrame
print(results_df)
# Plot the DataFrame as a table with increased font size
fig, ax = plt.subplots(figsize=(14, 8))  
ax.axis('tight')
ax.axis('off')
table = ax.table(cellText=results_df.values, colLabels=results_df.columns, loc='center')
# Adjust font size
table.auto_set_font_size(False)
table.set_fontsize(10)  
# Save the table as an image
plt.savefig('performance_metrics_table.jpg')
plt.show()

                 Model Accuracy % Precision % Recall % F1-Score
0  Logistic Regression     82.30%      85.43%   77.89%     0.81
1        Random Forest     82.77%      82.17%   83.70%     0.83
2                  KNN     76.13%      75.07%   78.25%     0.77
3          Naive Bayes     79.83%      80.71%   78.39%     0.80


# Define data
models = ['Logistic Regression', 'Random Forest', 'KNN', 'Naive Bayes']
y_true = [Y_test, Y_test, Y_test, Y_test]  # Replace with your true labels
y_pred = [y_pred_best, y_pred_best_rf, best_k_predictions, optimised_gnb_Y_predict]  # Replace with your predicted labels
# Calculate precision, recall, and F1-score
precision_scores = [precision_score(true, pred) for true, pred in zip(y_true, y_pred)]
recall_scores = [recall_score(true, pred) for true, pred in zip(y_true, y_pred)]
f1_scores = [f1_score(true, pred) for true, pred in zip(y_true, y_pred)]
# Set the width of the bars
bar_width = 0.25
# Set the positions of the bars on the x-axis
r1 = np.arange(len(models))
r2 = [x + bar_width for x in r1]
r3 = [x + bar_width for x in r2]
# Plot grouped bar chart with custom colors
plt.bar(r1, precision_scores, color='skyblue', width=bar_width, edgecolor='grey', label='Precision')
plt.bar(r2, recall_scores, color='green', width=bar_width, edgecolor='grey', label='Recall')
plt.bar(r3, f1_scores, color='red', width=bar_width, edgecolor='grey', label='F1-Score')
# Add xticks on the middle of the group bars
plt.xlabel('Models', fontweight='bold')
plt.xticks([r + bar_width for r in range(len(models))], models, rotation=45)
# Add labels and title
plt.ylabel('Scores', fontweight='bold')
plt.title('Precision, Recall, and F1-Score by Model')
# Add legend
plt.legend()
# Show plot
plt.tight_layout()
plt.savefig('PRF.jpg')
plt.show()

	Customer ID	Gender	Age	Married	Number of Dependents	City	Zip Code	Latitude	Longitude	Number of Referrals	...	Payment Method	Monthly Charge	Total Charges	Total Refunds	Total Extra Data Charges	Total Long Distance Charges	Total Revenue	Customer Status	Churn Category	Churn Reason
0	0002-ORFBO	Female	37	Yes	0	Frazier Park	93225	34.827662	-118.999073	2	...	Credit Card	65.60	593.30	0.00	0	381.51	974.81	Stayed	NaN	NaN
1	0003-MKNFE	Male	46	No	0	Glendale	91206	34.162515	-118.203869	0	...	Credit Card	-4.00	542.40	38.33	10	96.21	610.28	Stayed	NaN	NaN
2	0004-TLHLJ	Male	50	No	0	Costa Mesa	92627	33.645672	-117.922613	0	...	Bank Withdrawal	73.90	280.85	0.00	0	134.60	415.45	Churned	Competitor	Competitor had better devices
3	0011-IGKFF	Male	78	Yes	0	Martinez	94553	38.014457	-122.115432	1	...	Bank Withdrawal	98.00	1237.85	0.00	0	361.66	1599.51	Churned	Dissatisfaction	Product dissatisfaction
4	0013-EXCHZ	Female	75	Yes	0	Camarillo	93010	34.227846	-119.079903	3	...	Credit Card	83.90	267.40	0.00	0	22.14	289.54	Churned	Dissatisfaction	Network reliability
5	0013-MHZWF	Female	23	No	3	Midpines	95345	37.581496	-119.972762	0	...	Credit Card	69.40	571.45	0.00	0	150.93	722.38	Stayed	NaN	NaN
6	0013-SMEOE	Female	67	Yes	0	Lompoc	93437	34.757477	-120.550507	1	...	Bank Withdrawal	109.70	7904.25	0.00	0	707.16	8611.41	Stayed	NaN	NaN
7	0014-BMAQU	Male	52	Yes	0	Napa	94558	38.489789	-122.270110	8	...	Credit Card	84.65	5377.80	0.00	20	816.48	6214.28	Stayed	NaN	NaN
8	0015-UOCOJ	Female	68	No	0	Simi Valley	93063	34.296813	-118.685703	0	...	Bank Withdrawal	48.20	340.35	0.00	0	73.71	414.06	Stayed	NaN	NaN
9	0016-QLJIS	Female	43	Yes	1	Sheridan	95681	38.984756	-121.345074	3	...	Credit Card	90.45	5957.90	0.00	0	1849.90	7807.80	Stayed	NaN	NaN

	Age	Number of Dependents	Zip Code	Latitude	Longitude	Number of Referrals	Tenure in Months	Avg Monthly Long Distance Charges	Avg Monthly GB Download	Monthly Charge	Total Charges	Total Refunds	Total Extra Data Charges	Total Long Distance Charges	Total Revenue
count	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	6361.000000	5517.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000
mean	46.509726	0.468692	93486.070567	36.197455	-119.756684	1.951867	32.386767	25.420517	26.189958	63.596131	2280.381264	1.962182	6.860713	749.099262	3034.379056
std	16.750352	0.962802	1856.767505	2.468929	2.154425	3.001199	24.542061	14.200374	19.586585	31.204743	2266.220462	7.902614	25.104978	846.660055	2865.204542
min	19.000000	0.000000	90001.000000	32.555828	-124.301372	0.000000	1.000000	1.010000	2.000000	-10.000000	18.800000	0.000000	0.000000	0.000000	21.360000
25%	32.000000	0.000000	92101.000000	33.990646	-121.788090	0.000000	9.000000	13.050000	13.000000	30.400000	400.150000	0.000000	0.000000	70.545000	605.610000
50%	46.000000	0.000000	93518.000000	36.205465	-119.595293	0.000000	29.000000	25.690000	21.000000	70.050000	1394.550000	0.000000	0.000000	401.440000	2108.640000
75%	60.000000	0.000000	95329.000000	38.161321	-117.969795	3.000000	55.000000	37.680000	30.000000	89.750000	3786.600000	0.000000	0.000000	1191.100000	4801.145000
max	80.000000	9.000000	96150.000000	41.962127	-114.192901	11.000000	72.000000	49.990000	85.000000	118.750000	8684.800000	49.790000	150.000000	3564.720000	11979.340000

	Customer ID	Gender	Age	Married	Number of Dependents	City	Zip Code	Latitude	Longitude	Number of Referrals	...	Payment Method	Monthly Charge	Total Charges	Total Extra Data Charges	Total Long Distance Charges	Total Revenue	Customer Status	Churn Category	Churn Reason
7031	9974-JFBHQ	Male	31	No	1	Del Rey	93616	36.657462	-119.595293	0	...	Credit Card	99.15	6171.20	0	1434.24	7605.44	Stayed	NaN	NaN
7032	9975-GPKZU	Male	62	Yes	3	Alameda	94501	37.774633	-122.274434	0	...	Credit Card	19.75	856.50	0	352.82	1209.32	Stayed	NaN	NaN
7034	9978-HYCIN	Male	72	Yes	1	Bakersfield	93301	35.383937	-119.020428	1	...	Bank Withdrawal	84.95	4018.05	80	1987.63	6085.68	Stayed	NaN	NaN
7035	9979-RGMZT	Female	20	No	0	Los Angeles	90022	34.023810	-118.156582	0	...	Credit Card	94.05	633.45	0	255.43	888.88	Stayed	NaN	NaN
7036	9985-MWVIX	Female	53	No	0	Hume	93628	36.807595	-118.901544	0	...	Credit Card	70.15	70.15	0	42.09	112.24	Churned	Competitor	Competitor had better devices
7037	9986-BONCE	Female	36	No	0	Fallbrook	92028	33.362575	-117.299644	0	...	Bank Withdrawal	20.95	85.50	0	8.04	93.54	Churned	Competitor	Competitor made better offer
7038	9987-LUTYD	Female	20	No	0	La Mesa	91941	32.759327	-116.997260	0	...	Credit Card	55.15	742.90	0	606.84	1349.74	Stayed	NaN	NaN
7039	9992-RRAMN	Male	40	Yes	0	Riverbank	95367	37.734971	-120.954271	1	...	Bank Withdrawal	85.10	1873.70	0	356.40	2230.10	Churned	Dissatisfaction	Product dissatisfaction
7041	9993-LHIEB	Male	21	Yes	0	Solana Beach	92075	33.001813	-117.263628	5	...	Credit Card	67.85	4627.65	0	142.04	4769.69	Stayed	NaN	NaN
7042	9995-HOTOH	Male	36	Yes	0	Sierra City	96125	39.600599	-120.636358	1	...	Bank Withdrawal	59.00	3707.60	0	0.00	3707.60	Stayed	NaN	NaN

	Age	Tenure in Months	Avg Monthly Long Distance Charges	Avg Monthly GB Download	Monthly Charge	Total Charges	Total Refunds	Total Long Distance Charges	Total Revenue
0	37	9	42.39	16.0	65.6	593.30	0.00	381.51	974.81
1	46	9	10.69	10.0	-4.0	542.40	38.33	96.21	610.28
2	50	4	33.65	30.0	73.9	280.85	0.00	134.60	415.45
3	78	13	27.82	4.0	98.0	1237.85	0.00	361.66	1599.51
4	75	3	7.38	11.0	83.9	267.40	0.00	22.14	289.54

Telecom Customer Churn Prediction¶

Library Importation¶

Data Importation¶

Initial Data Analysis¶

Removing Joined from Target Variables¶

Exploratory Data Analysis(EDA) / Data Visualisation¶

Univariate Analysis¶

Dividing features into Categorical and Numerical for easy Visualisation¶

Bi-Variate Analysis¶

Visualising relationship between Categorical features and target variable¶

Data Cleaning¶

Removing Unnecessary Columns¶

Checking for Null in Columns and replacing with median and mode¶

Encoding Categorical Features¶

Normlisation¶

Correlation Analysis for Multi-Collinearity Assessment¶

Variance Inflation Factor (VIF) for Numerical Features¶

Chi Square Test for Categorical Features¶

Data Splitting for Machine Learning¶

Seperating Features and Target Variables¶

Backward Stepwise Regression¶

Data Splitting¶

Target Variable Distribution & Upsampling¶

Plotting Customer Status before upsampling¶

Upsampling using SMOTE¶

Machine Learning Algorithms¶

Logistic Regression¶

Optimisation of Logistic Regression Model¶

Random Forest Classifier¶

Optimising the Random Forest Model¶

K Nearest Neighbour¶

Optimising the KNN Model¶

Guassian Naives Bayes (GNB) Model¶

Optimising the GNB Model¶

Machine Learning Models Comparison¶

Accuracy Comparison¶

Receiver Operating Characteristic (ROC) - Comparison of Models¶

Precision-Recall & F1 Comparison¶

	Gender	Married	City	Offer	Phone Service	Multiple Lines	Internet Service	Internet Type	Online Security	...	Streaming Movies	Streaming Music	Unlimited Data	Contract	Total Extra Data Charges	Paperless Billing	Payment Method	Churn Category	Churn Reason	Customer Status
0	Female	Yes	Frazier Park	NaN	Yes	No	Yes	Cable	No	...	No	No	Yes	One Year	0	Yes	Credit Card	NaN	NaN	Stayed
1	Male	No	Glendale	NaN	Yes	Yes	Yes	Cable	No	...	Yes	Yes	No	Month-to-Month	10	No	Credit Card	NaN	NaN	Stayed
2	Male	No	Costa Mesa	Offer E	Yes	No	Yes	Fiber Optic	No	...	No	No	Yes	Month-to-Month	0	Yes	Bank Withdrawal	Competitor	Competitor had better devices	Churned
3	Male	Yes	Martinez	Offer D	Yes	No	Yes	Fiber Optic	No	...	Yes	No	Yes	Month-to-Month	0	Yes	Bank Withdrawal	Dissatisfaction	Product dissatisfaction	Churned
4	Female	Yes	Camarillo	NaN	Yes	No	Yes	Fiber Optic	No	...	No	No	Yes	Month-to-Month	0	Yes	Credit Card	Dissatisfaction	Network reliability	Churned

	Gender	Age	Married	Number of Dependents	Number of Referrals	Tenure in Months	Phone Service	Avg Monthly Long Distance Charges	Multiple Lines	Internet Service	...	Customer Status	Internet Type_Cable	Internet Type_DSL	Internet Type_Fiber Optic	Contract_Month-to-Month	Contract_One Year	Contract_Two Year	Payment Method_Bank Withdrawal	Payment Method_Credit Card	Payment Method_Mailed Check
0	0	37	1	0	2	9	1	42.39	0	1	...	1	True	False	False	False	True	False	False	True	False
1	1	46	0	0	0	9	1	10.69	1	1	...	1	True	False	False	True	False	False	False	True	False
2	1	50	0	0	0	4	1	33.65	0	1	...	0	False	False	True	True	False	False	True	False	False
3	1	78	1	0	1	13	1	27.82	0	1	...	0	False	False	True	True	False	False	True	False	False
4	0	75	1	0	3	3	1	7.38	0	1	...	0	False	False	True	True	False	False	False	True	False
5	0	23	0	3	0	9	1	16.77	0	1	...	1	True	False	False	True	False	False	False	True	False
6	0	67	1	0	1	71	1	9.96	0	1	...	1	False	False	True	False	False	True	True	False	False
7	1	52	1	0	8	63	1	12.96	1	1	...	1	False	False	True	False	False	True	False	True	False
8	0	68	0	0	0	7	1	10.53	0	1	...	1	False	True	False	False	False	True	True	False	False
9	0	43	1	1	3	65	1	28.46	1	1	...	1	True	False	False	False	False	True	False	True	False

	Gender	Age	Married	Number of Dependents	Number of Referrals	Tenure in Months	Phone Service	Avg Monthly Long Distance Charges	Multiple Lines	Internet Service	...	Customer Status	Internet Type_Cable	Internet Type_DSL	Internet Type_Fiber Optic	Contract_Month-to-Month	Contract_One Year	Contract_Two Year	Payment Method_Bank Withdrawal	Payment Method_Credit Card
0	0	0.295082	1	0.000000	0.181818	0.112676	1	0.844835	0	1	...	1	1	0	0	0	1	0	0	1
1	1	0.442623	0	0.000000	0.000000	0.112676	1	0.197632	1	1	...	1	1	0	0	1	0	0	0	1
2	1	0.508197	0	0.000000	0.000000	0.042254	1	0.666394	0	1	...	0	0	0	1	1	0	0	1	0
3	1	0.967213	1	0.000000	0.090909	0.169014	1	0.547366	0	1	...	0	0	0	1	1	0	0	1	0
4	0	0.918033	1	0.000000	0.272727	0.028169	1	0.130053	0	1	...	0	0	0	1	1	0	0	0	1
5	0	0.065574	0	0.333333	0.000000	0.112676	1	0.321764	0	1	...	1	1	0	0	1	0	0	0	1
6	0	0.786885	1	0.000000	0.090909	0.985915	1	0.182728	0	1	...	1	0	0	1	0	0	1	1	0
7	1	0.540984	1	0.000000	0.727273	0.873239	1	0.243977	1	1	...	1	0	0	1	0	0	1	0	1
8	0	0.803279	0	0.000000	0.000000	0.084507	1	0.194365	0	1	...	1	0	1	0	0	0	1	1	0
9	0	0.393443	1	0.111111	0.272727	0.901408	1	0.560433	1	1	...	1	1	0	0	0	0	1	0	1

	Age	Married	Number of Dependents	Number of Referrals	Avg Monthly Long Distance Charges	Avg Monthly GB Download	Online Security	Online Backup	Device Protection Plan	Premium Tech Support	Streaming TV	Streaming Movies	Paperless Billing	Total Refunds	Contract_Month-to-Month	Contract_One Year	Contract_Two Year
2059	0.606557	0	0.000000	0.000000	0.111474	21.0	0	0	0	0	0	0	1	0.00	1	0	0
4323	0.491803	1	0.000000	0.909091	0.903838	21.0	0	0	0	0	1	1	1	0.00	0	1	0
1544	0.163934	1	0.222222	0.909091	0.504492	69.0	1	1	1	1	0	0	0	0.00	0	0	1
2644	0.770492	1	0.000000	0.545455	0.634341	8.0	1	0	0	0	1	1	1	0.00	0	1	0
4029	0.655738	1	0.111111	1.000000	0.841364	25.0	1	0	1	1	0	0	1	25.49	0	1	0